SDE vs EBITDA: Which Number Your Home-Services Business Sells On

If you’re pricing a home-services business, the multiple is the second question. The first is which earnings number the multiple applies to.

Seller’s discretionary earnings (SDE) is your pre-tax profit plus everything one working owner takes out of the business: your salary and its payroll taxes, your health insurance, the truck you drive home, plus interest, depreciation, and documented one-time expenses. It answers: how much does this business put in one owner-operator’s pocket per year?

EBITDA is earnings before interest, taxes, depreciation, and amortization, with owner compensation left in as a normal operating expense at market rate. It answers: what does this business earn for whoever owns it, assuming someone has to be paid to run it?

Same company, two correct numbers, often $100,000 apart. Below: both calculations on one realistic $1.8M P&L, the point where buyers stop using one and start using the other, and the add-backs that survive diligence versus the ones that torch your credibility.

The two definitions, one level deeper

SDE starts from pre-tax net income and adds back:

• Owner’s total compensation (salary, payroll taxes, health insurance, retirement)
• Interest, depreciation, and amortization
• Personal expenses run through the business
• Genuinely one-time expenses, with documentation

One important limit: SDE assumes one full-time working owner. If you and a second owner both work the business full time, only one compensation package gets added back. The other gets replaced at a market-rate salary, because the buyer has to pay someone to do that job.

EBITDA adds back interest, taxes, depreciation, and amortization, and stops there. In a sale, what actually gets quoted is adjusted EBITDA: the same personal-expense cleanups apply, but owner pay gets normalized to what a general manager would cost, not added back entirely.

The practical bridge between the two:

Adjusted EBITDA is roughly SDE minus the market cost of replacing you.

That one sentence explains most of the confusion between broker listings and PE letters of intent.

Worked example: one $1.8M shop, both ways

Take a coatings and install business doing $1.8M in revenue. The owner runs sales and scheduling and takes a $120K salary. His spouse is on payroll at $30K with no documented role. One truck doubles as his personal vehicle. Last year the company settled a one-time lawsuit.

Here’s the SDE build:

Now the same shop on an adjusted EBITDA basis. Every cleanup above still applies. The difference is one line: instead of adding back the owner’s full package, the buyer charges the business what a general manager costs. For a $1.8M home-services operation, call it $95,000 plus about $7,000 in payroll taxes.

SDE comes out 29% higher than adjusted EBITDA on the identical business. Neither number is wrong. They answer different questions.

Why mixing them up costs real money

Multiples are quoted against a specific earnings basis, and applying one basis’s multiple to the other basis’s number produces garbage:

• 2.8x SDE on this shop is about $1.27M. A buyer quoting 3.6x adjusted EBITDA lands at about $1.26M. Different labels, nearly the same price.
• An owner who hears “5x” on a private equity podcast and applies it to his SDE thinks his shop is worth $2.27M. It isn’t.
• A buyer who applies a 2.8x SDE-style multiple to your EBITDA number is quietly lowballing you by about $285,000.

When anyone quotes you a multiple, your first question is: of which number?

When buyers stop using SDE and switch to EBITDA

The switch isn’t about revenue. It’s about whether the business still needs you in it.

MidStreet, a lower-middle-market M&A firm, publishes the crossover in earnings terms:

Broker-published ranges. Last verified 2026-07-28.

The logic: an individual buyer is replacing you personally, so the number that matters is the total benefit one working owner receives. That’s SDE. A private equity firm is not going to run your crews. It has to hire a manager, so it prices the business with that salary already deducted. That’s EBITDA.

This is also why crossing the threshold changes the multiple, not just the metric. Once your shop earns enough with a management layer in place, you graduate from the individual-buyer pool into the institutional pool, and institutional buyers pay higher multiples of a smaller number. Whether that nets out in your favor depends on the business. We ran that math in what home-services businesses actually sell for.

Add-backs that count, add-backs buyers reject

The whole gap between net income and SDE is add-backs, so this is where deals are won and lost. CT Acquisitions’ 2026 add-back guide draws the lines most brokers and quality-of-earnings analysts follow:

Two distinctions from that guide worth memorizing:

The family payroll line is about work, not family. A spouse paid $30,000 with no documented role is a clean add-back. The same spouse doing genuine bookkeeping at a market rate is not, because the buyer inherits that job and has to pay someone for it.

“One-time” gets tested against history. Quality-of-earnings analysts check the claim against three years of statements. An expense that appears in two of the last three years gets reclassified as recurring, and the add-back dies.

The gotchas that kill trust in diligence

Aggressive add-backs cost more than they add. A rejected add-back doesn’t return you to neutral. The buyer removes the dollar, the multiple removes it several times over, and now every other line you presented gets re-checked by someone who assumes you were optimistic everywhere. A conservative SDE that holds up beats an inflated one that gets negotiated down in front of a newly skeptical buyer.

Cash jobs you can’t prove don’t count. Revenue that never hit the tax return or the bank account does not exist for valuation purposes. Worse, telling a buyer about it is telling them your books lie, and they will price that honesty problem into everything else.

The metric switch can sneak up on you. Owners near the threshold sometimes market the business on SDE and then face institutional buyers who re-cut everything to EBITDA mid-process. If your earnings are anywhere near $1M, run both numbers before you go to market so nobody surprises you with your own P&L.

Whichever number you sell on, it has to be provable

Here’s the part the valuation guides skip: every add-back in that table is a documentation problem before it’s an accounting problem.

The mileage log for the truck. The engagement letter behind the legal fee. The three years of clean statements that prove the roof repair really was one-time. Job-level margins that tie back to invoices instead of a spreadsheet someone updates on Sunday nights.

This is where owner-operators lose money in diligence. Not because the earnings aren’t real, but because the records live in four places: the CRM says one thing, QuickBooks says another, job costing is a spreadsheet, and the export a buyer’s analyst asks for takes three weeks and doesn’t reconcile. Every gap turns an add-back from a fact into an argument, and arguments get settled at the buyer’s number.

That’s the operational side we work on. We build reporting and job-costing systems on top of Jobber, Housecall Pro, and ServiceTitan for exactly this reason: so revenue, job costs, and margins tie out from the CRM to the books, and the number you claim is the number your records show. The owners who start that cleanup two years before a sale walk into diligence with answers instead of explanations.

Run both numbers, then pick your lane

The takeaway in three lines:

• Under roughly $1M in earnings: your number is SDE. Compute it honestly, document every add-back, and expect 2 to 3x from individual buyers.
• Above roughly $1.5M with a management layer: your number is EBITDA, your buyers are institutional, and the multiple range moves up.
• In between: run both, because you’ll hear offers quoted both ways, and the basis matters more than the multiple.

For what those multiples translate to in actual dollars by trade and size, read what home-services businesses actually sell for. And if your add-backs currently live in a spreadsheet and a shoebox, that’s a fixable problem, and fixing it before a buyer shows up is worth real money.